The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The CI/CD integration guide in references/ci-cd-integration.md includes a workflow that downloads the rudder-cli binary from the vendor's official GitHub repository. This is a standard operation for setting up the required tooling.
Evidence: curl -fsSLO "https://github.com/rudderlabs/rudder-iac/releases/download/$tag/rudder-cli_Linux_x86_64.tar.gz" in references/ci-cd-integration.md.
Security Practice: The installation script implements a security best practice by verifying the downloaded archive against an official checksum file using sha256sum before extraction.
[INDIRECT_PROMPT_INJECTION]: The skill processes external tracking plan data (YAML) to generate code, which is a potential surface for indirect injection if the source data is untrusted.
Ingestion points: Local files like tracking-plan.yaml and ruddertyper.yml as described in SKILL.md.
Boundary markers: Not explicitly defined in the instructions, though the rudder-cli tool performs validation on the input schema.
Capability inventory: The skill uses rudder-cli to validate plans, apply changes to a remote workspace, and generate source code (Swift/Kotlin). It also utilizes standard file read/write/edit tools.
Sanitization: Relies on the internal validation logic of the rudder-cli tool; the skill itself does not specify additional sanitization steps for the input data.

rudder-typer-workflow